Measuring patients’ medical treatment preferences in advance care planning : development and validation of the Treat-Me-ACP instrument - a secondary analysis of a cluster-randomized controlled trial

Abstract

Background:

Advance Care Planning interventions should be evaluated as broadly as possible to gain a holistic understanding of the Advance Care Planning process. However, validated early stage outcome instruments are lacking. Therefore, the Treatment-Preference-Measure-Advance Care Planning (Treat-Me-ACP) instrument was developed and validated as part of the cluster-randomized controlled trial STADPLAN (Study on Advance Care Planning in care-dependent community-dwelling older persons) to assess the effects of Advance Care Planning interventions on patients’ medical treatment preferences. Methods:

The design of Treat-Me-ACP is based on the Emanuel Medical Directive and the Life Support Preferences Questionnaires. Using a multi-stage team approach a preliminary version of the Treat-Me-ACP was developed and pre-tested. The pre-tested instrument consists of one global medical care goal-item, five hypothetical scenarios with five hypothetical treatments, and one how would you feel-item within each scenario. A total of five scenario preference scores and five treatment preference scores can be formed. This version was subsequently applied to a subsample of the STADPLAN project (n = 80) to assess patient’s preferences at baseline (T0) and at 12-month follow-up (T2). The further validation steps were based on this subsample and included: (1) acceptance by using completion rate and frequencies of missing data, (2) internal consistency by using Cronbach’s α to test whether it was possible to create preference scores by scenario and treatment, (3) concurrent validation examining the association between the global medical care goal-item and the preference scores and the association between the how would you feel-items and the scenario preference scores, and (4) responsiveness of the instrument to changes in preferences for life-sustaining treatments by comparing preference scores from T0 to T2 between study groups. Results:

Acceptance of the instrument was high. Results of concurrent validation indicate that the five scenarios represent the global medical care goal well. The preference scores showed an average tendency for decreasing preferences for life-sustaining treatments across all scales for the intervention group during study follow-up. Conclusions:

The Treat-Me-ACP can be used to evaluate the dynamics of patients’ medical treatment preferences in Advance Care Planning. It has been validated for care-dependent community-dwelling older persons and can be used as an additional outcome measure in evaluating the effectiveness of ACP interventions.